| HN Mirror

"This article is very painfully committing the Base Rate Fallacy."

You didn't read past the first sentence, did you? Embarrassing, because the article starts by explaining the base rate fallacy and pointing out that the social media meme it starts by highlighting is wrong. Then it goes on to do correct analysis, which shows the conclusions I gave. You'd have known this if you read my post properly too, because I have a whole paragraph about the stats involved in correctly calculating base rates. If you think there are mistakes in the rest of the article please explain them, but you do need to actually read it first.

"I'm not going to accept "Daily Sceptic" as a source."

Again you didn't even click the link and look, did you? The image is a table of data from official government statistics, they simply happen to host the screenshot. But I knew I'd get a response like that from someone. Lack of intellectual curiosity around this topic is extreme - for obvious and understandable reasons of course. But still.

Re: PowerBI

Where do you see that? There are virtually no COVID deaths (a.k.a. "had a positive test a month before death") since the end of the winter in the UK. Look at the data for the various kinds of heart failure, for example. You see clear excess where COVID isn't implicated starting around the end of April.

"1 additional death means essentially zero correlation."

After incorrectly snarking about not understanding statistics, you're now demonstrating a mis-understanding yourself (albeit a very common one).

You can't simply look at a small difference and say "not statistically significant therefore there is no risk". That's not how statistics works. Firstly, the overall sample size was very large, it was an RCT. So we can say with great certainty that the vaccines have no effect on mortality, yet, that was the entire purpose of developing them. I see up thread some people are now trying to deny this, claiming that the vaccine trials were never meant to even study death rates! Truly Orwellian stuff. Death is the endpoint that motivated everything.

What we can't say with great certainty is if the vaccines are truly more deadly than the placebo. But statistical significance is not the same thing as significance. This is a really common logic error you see even amongst scientists themselves (when badly motivated). This result means the vaccines might be more deadly than the placebo or might not, and therefore the correct response is to gather more data. The incorrect answer is to say "eh, yolo, let's assume the optimistic result", especially if you're about to force people to take it on a massive scale.

But of course they didn't gather this data. The people who created and run the COVID vaccine programmes think that any expression of doubt about vaccine safety is immoral anti-science anti-vaxx insanity, which in turn means they can't neutrally measure or act on data. Their conclusions are chosen before they even do a single experiment, so they just went ahead and did it.

At any rate, the placebo in these trials was incorrect. They gave the placebo arm vaccines too, just different ones, so this is actually not comparing against reality (=no vaccine) and therefore overly generous to the vaccine under test.